Uncertainty-Aware Flexibility Requirements Disaggregation
Michael Lechl, Hermann de Meer · 2025
Virtual power plants have emerged as a result of the energy transition and tapping new types of flexibility resources at household and community levels.Virtual power plants offer flexibility services to other system actors by aggregating the flexibility potentials of flexibility resources such as battery storage systems.When a service is requested, for example to balance fluctuations in renewable generation, a virtual power plant must decide on the share of each flexibility resource in fulfilling the request.This work addresses the problem of disaggregating service requests, i.e., flexibility requirements, among flexibility resources by proposing an evolutionary algorithm.The main contribution is integrating stochastic flexibility calculus with the evolutionary algorithm to enable uncertainty of flexibility potentials, particularly on the demand side, to be considered.Uncertainty creates a trade-off between minimizing the risk of lacking flexibility potentials and avoiding wasting (demand-side) flexibility potentials.To address this trade-off during disaggregation, an indicator for assessing the probability of wasting flexibility potential is introduced as an extension of the stochastic flexibility calculus.An evaluation for an example setup shows the effectiveness of the evolutionary algorithm in reducing the risk of lacking flexibility potentials while avoiding waste of flexibility potentials.